From: Tom Hughes (thh_at_cyberscience.com)
Date: Mon Jan 14 2002 - 08:31:50 GMT
In message <200201111853.KAA28214_at_hunter>
Rod Armstrong <rod_at_san-jose.tt.slb.com> wrote:
> Tom Hughes suggestion of a usleep(10) in the ptrace_wait_for_target()
> loop is probably OK. I don't have a Redhat system, but clearly something
> changed in the wait3() call from Redhat 6.2 to Redhat 7+. For SunOS,
> which uses ptrace, wait() is used, which is blocking but interruptible.
One of my colleagues suggested that the difference might be that the
older 2.2 kernel needed to grab the master kernel lock to do the wait
which would have meant the task had a good chance of having to wait
while the 2.4 kernel has finer grained locking which probably makes a
wait much less likely.
> On Linux, wait3() must be used in order to get interrupts. Maybe
> someone with a Redhat system could investigate.
I don't see why wait() shouldn't get interrupts on linux... I will
take a look at it.
Tom
--
Tom Hughes (thh_at_cyberscience.com)
Software Engineer, Cyberscience Corporation
http://www.cyberscience.com/
This archive was generated by hypermail 2.1.4 : Wed Feb 13 2002 - 21:51:35 GMT